Reduction of test and maintenance human errors by analyzing task characteristics and work conditions

Human errors during test and maintenance activities had a significant contribution to unplanned reactor trips of nuclear power plants. Especially, four error modes including wrong object, omission, too little, and wrong action appeared to be recurrently occurring over decades for execution-based human errors. This paper introduces human error analysis (HEA) procedures for a predictive analysis of human error potentials when maintenance personnel perform test or maintenance actions based on a work procedure or work plan. The developed HEA procedures are based on the basic error characteristics and performance shaping factors (PSFs) associated with specific error modes, and each of the procedures is composed of three steps such as analysis of basic error potential, evaluation of possible impacts on the system, and identification of deficient work context or PSFs. The procedures can be used effectively for identifying potential human error modes leading to potential negative impacts such as unplanned reactor trips or plant transients and analyzing weak points of work context while performing T&M works in normal operating conditions, thereby the maintenance personnel can prepare for the identified human errors and work context to prevent or reduce potential human errors. A case study was illustrated in this paper.

► Task characteristics and performance shaping factors (PSFs) for specific error modes.
► Human error prediction for maintenance tasks based on error characteristics and PSFs.
► Human error procedures are effective for predicting error modes and weak conditions.
► Maintenance personnel can prepare against identified human errors and weak conditions.